g model article in press - uzh 144/2016_parajuly... · discarded weee in residual waste from danish...

R

F

WF

KSC

a

ARRAA

KWDUEDR

1

oedst5ypa

stdem

h0

ARTICLE IN PRESSG ModelECYCL-3325; No. of Pages 8

Resources, Conservation and Recycling xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Resources, Conservation and Recycling

journa l homepage: www.e lsev ier .com/ locate / resconrec

ull length article

aste electrical and electronic equipment (WEEE) in Denmark:lows, quantities and management

eshav Parajuly ∗, Komal Habib, Gang LiuDU Life Cycle Engineering, Department of Chemical Engineering, Biotechnology and Environmental Technology, University of Southern Denmark,ampusvej 55, 5230 Odense M, Denmark

r t i c l e i n f o

rticle history:eceived 11 April 2016eceived in revised form 4 August 2016ccepted 5 August 2016vailable online xxx

eywords:EEE

enmarkrban mininglectronic waste

a b s t r a c t

In this study, we comprehensively map and estimate the flows of electrical and electronic equipment(EEE) and the corresponding waste (WEEE) in Denmark. The quantitative analysis is supplemented witha thorough diagnosis of the WEEE management system. Dynamic material flow analysis (MFA) is used toestimate the flows for the period of 1990–2025. The estimates are based on sales data of 61 householdproducts – equivalent to 80% (by weight) of the total household EEE – and their lifespans modelled usingWeibull distribution function. Building on this, the potential resources available for recovery over time,and their corresponding revenues are evaluated. The results show that the amount of WEEE generatedper year increased from 45 to 81 kilo tons (kt) between 1990 and 2015. The amount of EEE put on market(PoM), on the other hand, peaked to 101 kt in 2006 from 61 kt in 1991, but declined to 84 kt in 2015. Interms of the PoM quantity, the EEE market is found saturated, and can be expected to remain largely

ynamic material flow analysisesource recovery

unchanged over the next decade. Consequently, there will not be any significant increase in WEEE quan-tities. Denmark has a well-established WEEE management system that has been performing adequatelyagainst the WEEE Directive. However, the new set of legislations means a need for recalibration of theperformance indicators for the system. A more robust and systematic documentation of the flows willsupport the WEEE management system in achieving higher resource recovery.

© 2016 Elsevier B.V. All rights reserved.

. Introduction

With the rapid technological advancement and economic devel-pment, the multiplicity and use of electrical and electronicquipment (EEE) have increased drastically over the last twoecades. Consequently, the amount of end-of-life (EoL) products,ynonymously known as electronic waste (e-waste) or waste elec-rical and electronic equipment (WEEE), is also on the rise. Nearly0 million metric tons (Mt) of WEEE is forecasted globally for theear 2018, which will be a 50% increase compared to 2010. Euro-ean countries alone produced 11.6 Mt of WEEE in 2014 with anverage of 15.6 kg per capita (Balde et al., 2015b).

While the increasing amount is a challenge for its managementystem, WEEE also offers a secondary source of valuable resourceshat are readily available for recycling. This not only reduces the

Please cite this article in press as: Parajuly, K., et al., Waste electrical

and management. Resour Conserv Recy (2016), http://dx.doi.org/10.1

emand for finite primary resources, but also brings economic andnvironmental benefits through avoided adverse impacts of pri-ary production (Habib et al., 2015; Mudd, 2010). In order to

∗ Corresponding author.E-mail address: [email protected] (K. Parajuly).

ttp://dx.doi.org/10.1016/j.resconrec.2016.08.004921-3449/© 2016 Elsevier B.V. All rights reserved.

plan effective management strategies, policy makers and EoL actorsneed a better understanding of the system, including informationon the amount of WEEE generation, material composition, and theiravailability for resource recovery (Zeng et al., 2015). Lack of insightinto such potentials can keep countries from implementing theright strategies.

In 2014, the amount of per capita put on market (PoM) EEEwas 21 kg in Denmark, the seventh highest among the 31 Euro-pean countries (DPA-System, 2015; Eurostat, 2016). Household EEEconsisted 77% of the total PoM amount, and more than 97% of thecollected WEEE was collected from households. As of 2015, an aver-age Danish household owns at least 27 common EEE items, witha total number of products in use being 33 million units acrossDenmark. Fig. 1 shows the increasing trend of household possessionof different EEE items in Denmark between 1990 and 2015.

As a member state of the European Union, Denmark has aWEEE management system that complies with the WEEE Direc-tive (European Parliament, 2003). Starting from August 2005, the

and electronic equipment (WEEE) in Denmark: Flows, quantities016/j.resconrec.2016.08.004

Directive required the member states to establish a separate col-lection and treatment system for household WEEE. Reporting ofthe quantities of PoM EEE, collected WEEE and their subsequenttreatment also became mandatory. The Danish Producer Respon-

dx.doi.org/10.1016/j.resconrec.2016.08.004


http://www.sciencedirect.com/science/journal/09213449

http://www.elsevier.com/locate/resconrec

mailto:[email protected]


ARTICLE IN PRESSG ModelRECYCL-3325; No. of Pages 8

2 K. Parajuly et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx

Table 1Process description.

Process Description

1. Producers Original equipment manufacturers and resellers of new products2. New products Put on market (PoM) Electrical and electronic equipment (EEE)3. Households Users of the PoM EEE, where EEE is stocked as

• In-use stock: EEE that are functional and in use• Hibernating stock: EEE that are functional, but not in use• Obsolete stock: EEE that are neither functional nor in use

4. Preparation for reuse Second-hand stores, refurbishment shops, and user-to-user resale platforms5. Official collection Collection facilitated by municipalities, collective schemes, retail stores, and producers6. Other collection routes EoL product collection outside the official system7. EoL products EEE discarded of by its user with no intention of using it again8. Preprocessor First stop in the WEEE treatment

9. Recyclates Outgoing material/component str10. End processor Final stop in the WEEE treatment

Fi

ssttdm(

DpEbitro(ntIbs

wheotiWbr

a

b

ig. 1. Trend of increasing use of electrical and electronic products in householdsn Denmark [adapted from (Statistics Denmark, 2016)].

ibility System (DPA-System, formerly known as WEEE-System)tarted reporting these data from 2006. These reports are based onhe data provided by the registered producers (PoM data) and byhe local authorities, who facilitate the collection (WEEE collectionata). Further, these reports provide some information on treat-ent of the collected WEEE as per the requirement of the Directive

DPA-System, 2015).The collection target of 4 kg per capita per year set by the

irective has been easily achieved in Denmark (with, for exam-le, 12.7 kg per capita collected in 2014). Although the data on PoMEE and collected WEEE are reported comprehensively, there haveeen no previous studies to estimate the actual WEEE generation

n Denmark. The Directive did not require any mandatory quan-ification at national level, and the collection target was fixed –egardless of the amount of WEEE generated. However, a numberf substantial changes were made in the recast of the DirectiveEuropean Parliament, 2012). The Directive (recast) requires theew collection targets to be based on the amount of PoM EEE ofhree preceding years or the WEEE generated in the same year.t implies that more robust record keeping and calculations wille required to document the performance of WEEE managementystem.

Bigum et al. (2013) have empirically studied the amount ofrongfully discarded WEEE in residual waste from Danish house-

olds. In another study, Habib et al. (2014) considered selectedlectronic products in order to quantify the secondary supplyf rare earth-based permanent magnets for Denmark. However,here are no previous studies on WEEE quantification consider-



ng a wide range of EEE items. Without the quantification of theEEE amounts, the actual potential of material recovery cannot

e properly understood. It also keeps both the authorities andecyclers from tracking the amounts leaking out of the official col-

chain (consists of dismantling, shredding & separation)eams resulting from the preprocessing of WEEEchain, where materials are recovered in their pure form

lection channels. Neither major mining activities nor treatmentfacilities for final recovery of plastics and metals from WEEE exist inDenmark. The current generalized preprocessing based on ‘shred-and-separate’ approach results in loss of potential revenues, whichis exported outside the country in the form of material recyclates(Parajuly et al., 2016). In this context, the knowledge of EEE andWEEE flows is important for a country like Denmark in order toexploit the potential – especially during collection and preprocess-ing – of available resources in WEEE.

In this study, we comprehensively map the flows and systemati-cally quantify WEEE generation in Denmark. Material flow analysis(MFA) is used to estimate the WEEE generation from householdproducts. MFA, which is based on the principle of material con-servation, is an established tool for decision support in waste andresource management (Brunner and Rechberger, 2004). DifferentMFA models have been used for quantifying WEEE at national lev-els (Duygan and Meylan, 2015; Kalmykova et al., 2015; Zeng et al.,2015; Zhang et al., 2012). Dynamic MFA models are used to under-stand the pathways, stocks, and flows of products and materialsover time (Muller et al., 2014). Such models are usually based onthe input (sales) data and the lifetime distribution function of dif-ferent products in a socio-economic system in order to quantify thedynamics and magnitude of EoL products (Zeng et al., 2015).

The MFA in this study is supplemented with a thorough diag-nosis of the WEEE management system and its performanceindicators. To our knowledge, this is the first analysis of WEEE gen-eration, recovery potentials, and management system in Denmark.We envision that this study will help the concerned stakeholdersto evaluate the existing policies, understand the implications of thefuture compliance, and support the transition towards meeting thenew set of legislative requirements.

The main objectives of this study are to:

) scientifically assess the quantities of WEEE generated in Danishhousehold and thus the availability of secondary resources;

) systematically describe the flows of products along their lifestages, including the flows not covered by the official channel(such as reuse and complementary recycling); and

c) identify the challenges and opportunities towards efficientresource recovery from EoL electronics against the changing leg-islations and market conditions.

2. Methodology

2.1. System definition


Our system definition in Fig. 2 shows all the processes and flowsrelevant to EEE and WEEE management system for Denmark. Thedescription of each process and flow is provided in Tables 1 and 2respectively.



K. Parajuly et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx 3

Fig. 2. Flow of products and embodied materials along their life s

Table 2Flow description.

Flows Description

A0-1 Resources used in EEEA1–2 New products entering EEE marketI2, E2 Import and export of new productsA2–3 PoM EEEA3–5 = A5–7 EoL products collected through the official collection systemA3–6 = A6–7 EoL products collected through complementary channelsA4–3 = A7–4 EoL products going for reuse in householdsI7, E7 Import and export of EoL products (as used products or WEEE)A7–8 WEEE entering the official recycling routeA7 WEEE processed outside the official systemA8–9 Outgoing material streams from preprocessing plantsA8 Material losses occurring during preprocessingI9, E9 Import and export of material recyclatesA Material streams going to end processing plants

2

lAycesa

2

wh(ntt

utkA

9–10

A10 Material losses occurring during end processingA10–0 Materials recovered as secondary resources

.2. Quantification of flows

After identifying all processes and flows, we quantitatively ana-yzed the PoM EEE (Flow A2–3) and the corresponding WEEE (Flows3–5 and A3–6) over the period of 1990–2025. This covers the pastears for which sales data were available, and the next decade toome (the duration of scenarios in the future is kept short consid-ring the uncertainties related to the development of EEE sector;ee Section 3.4 for details). The following provides sources of datand methodology used for this analysis.

.2.1. PoM EEEWe considered 61 different product types, for which sales data

ere readily available. This includes the most common house-old products from four categories: 1 – Large household appliancesLHA), 2 – Small household appliances (SHA), 3 – IT & telecommu-ication equipment (ITE), and 4 – Consumer electronics (CE). Allhe products are listed in Table S1 in Supplementary material (SM)ogether with their sales data.

Fig. 3 summarizes the time series for the quantitative analysis



sed in this study, and the sources and methods used to obtainhe relevant data. Primarily, the sales data reported by the mar-et research firm Euromonitor International (2016) were used.ccording to Euromonitor, the data are collected from different

tages within and outside the spatial boundary of Denmark.

sources including national statistics, retailers, trade associationsetc. For LHA, the sales data were also available from the DanishAssociation for Suppliers of Electrical Domestic Appliances (FEHA,2016) for the period of 1991–2012, which included 8 product typesfor 1991–2000, 10 types for 2001–2008, and 15 types for 2009onwards. We had a cross check and found these two sources tobe consistent with each other.

We considered the sales data between 1991 and 2015 to be thebaseline, and used linear extrapolation to backcast the historic salesdata for LHA to 1956. Assuming the maximum life of a product to be35 years, products sold in and after 1956 can be thus covered in thecalculation of WEEE generated from 1991 onwards. The sales dataof SHA before 2009, and ITE and CE before 2010 were not avail-able. Therefore, the historic sales data for these categories werecalculated based on LHA. For this, we assumed that the productsfrom other categories were sold in the same ratio to the productsin the category LHA. For the future projection, we used a no-growthscenario considering a saturated market with no significant demo-graphic and/or economic changes (which has been the trend in thepast decade not only in Denmark, but also in other similar countriessuch as Sweden, Norway, and Austria; see details in Fig. S2 in SM).The sales for each year from 2016 to 2025 are assumed to be theaverage of preceding six years.

The sales data provided the number of units of products. Theseunits were converted into the weight of PoM EEE by using the aver-age weight of each product type. The average weight of 44 out of the61 products were empirically determined, which builds on a sortingand dismantling experiment performed for the household WEEEcollected at the municipal collection points in Denmark (unpub-lished work). The remaining were adapted from Wang et al. (2012)and Zeng et al. (2015). Please refer to Table S2 in SM for the details.

2.2.2. WEEEWeibull distribution function has been widely used for defin-

ing the lifetime of products in order to estimate WEEE generation(Balde et al., 2015a; Elshkaki, 2005; Kalmykova et al., 2015; Polakand Drapalova, 2012; Zeng et al., 2015; Zhang et al., 2012). Thisapproach takes into account the dynamic nature of product obso-


lescence, contrary to the assumption of fixed product lifespan(Zhang et al., 2012). The lifespan profile (of the sales cohort) isdefined by the shape and scale parameters (� and �, respectively).The maximum lifetime of a product is considered to be 35 years,




Fig. 3. Sources and methods used to obtain the data for the different categories over timefor the number of products specified within [square brackets] and their sources within (data were backcasted and forecasted, respectively.

ws

vauwpf

te

wfrom 15 to 10% during the same period.

Fig. 4. Probability density function of failure rate for the four categories.

ithin which the probability of all products coming to their EoLums up to 1 for the given � and � values.

Balde et al. (2015a) have made available these parameters forarious products in the Netherlands, France, and Belgium. Wedapted the average � and � values at category level for the prod-cts from four categories: LHA, SHA, ITE, and CE. The failure ratesere calculated in Microsoft Excel 2010 using the shape and scale

arameters. Fig. 4 shows the resulting probability distribution andailure rates for the four categories.

WEEE generation for each year starting from 1957 to 2025 washen calculated using these failure rates. The amount of WEEE gen-rated for each year is given by the following equation:

WEEE(y) =∑4

i=1Pi(1956) ∗ fi(y−1956) + Pi(1957) ∗ fi(y−1957) + ... . ..

+Pi(y−1) ∗ fi(y−1), (1)

hereWEEE(y) = amount of WEEE generated in year y,i = category of WEEE (LHA, SHA, ITE, and CE),



Pi(y) = amount of category i EEE put-on-market in year y, andf = failure rate.

. The solid lines indicate the readily available sales data in the corresponding yearsround brackets). The short- and long- dotted lines show the period for which the

2.2.3. Secondary resourcesBuilding on the estimated amounts of WEEE generation, we

quantified the amount of secondary resources available for recy-cling. The material composition for this calculation is partly basedon our empirical study and partly adapted from Oguchi et al. (2011).When the material composition of the exact product type was notavailable, the composition of another similar product was used.The material compositions of individual products were then aggre-gated to category level and the secondary resources available werecalculated for each category. For this conversion from product tocategory material composition, weighted average was used basedon the product sales data for 2015. The secondary resource availablein the WEEE generated each year is given by:

Sjy =∑4

i=1WEEEi(y) ∗ cij, (2)

whereSjy = resource j available as WEEE in year y,WEEEi(y) = amount of Category i WEEE in year y, andcij = content of resource j in Category i WEEE.Finally, the potential revenues were calculated using the mate-

rial composition and their market price adapted from Cucchiellaet al. (2015) and our empirical study. See Table S3 in SM for thedetails.

3. Results and discussion

3.1. PoM EEE and WEEE quantities

The amounts of PoM EEE and the corresponding WEEE gener-ation were quantified at the category level for LHA, SHA, ITE, andCE. Fig. 5 shows the estimated amounts for each category for theperiod between 1990 and 2025. As it can be seen, the amount ofthe PoM EEE peaked in the year 2006 with a total of 101 kt, growingfrom 60 kt in 1990. In the following years, however, it declined toreach 84 kt in 2015, and it is not expected to change significantly innear future. The share of each category to the total amount of EEEhas remained consistent with little change over time, with LHAbeing the most significant part. During the period of 2009–2015,for which complete sales data were available, the share of LHAincreased from 70 to 74%, while SHA and ITE both increased by1% to reach 7 and 10% respectively. The weight share of CE declined


Our estimates show that the annual WEEE generation increasedfrom 45 to 81 kt between 1990 and 2015 with a growth rate of 2–3%each year until 2011. However, this growth started slowing down


ARTICLE ING ModelRECYCL-3325; No. of Pages 8

K. Parajuly et al. / Resources, Conservation

Fa

fnoato

wa(psToApi

op2Wefguh

3

asgFB

cF3ta2ejBtp

period of 2006–2014 (see SM, Table S4). According to the Directive

ig. 5. Quantity of new products put on market (bars) and the EoL products (curvedrea) for 1990–2025.

rom 2012, and the WEEE generation is projected not to change sig-ificantly over the next decade. It is important to mention here thatur estimates are based on only 61 products and thus do not coverll EEE types. Nevertheless, compared to the PoM data reported byhe DPA-system for 2006–2014, our estimates on average cover 80%f total household EEE. See Section 3.4 for the details.

We compared the trend of EEE and WEEE generation in Denmarkith two Scandinavian neighbors (Norway and Sweden) and

nother European country with similar socio-economic featuresAustria). The amounts of PoM EEE and collected WEEE for theeriod of 2006–2013 in all four countries have been largely con-istent with virtually no growth after 2008 (see Fig. S2 in SM).his trend suggests a saturated EEE market as well as the presencef well-established WEEE collection systems in these countries.ssuming no significant change in population growth or individualurchasing power, this trend can be expected to remain constant

n the coming years.In countries with increasing population and economic devel-

pment, the outlook appears very different. In China, for example,er capita WEEE generation is forecasted to increase from 6 kg in014 to 18 kg in 2030 (Zeng et al., 2015). In India, the amount ofEEE increased from 146 kt in 2005 to 800 kt in 2012 (Sharma

t al., 2016). In Brazil, the per capita WEEE generation increasedrom 3.8 kg in 2008 to 7 kg in 2014 (de Souza et al., 2016). Therowth rates in these major developing countries indicate a contin-ing increase in global amount of WEEE. Countries like Denmark,owever, will see a little change.

.2. Secondary resources and potential revenue

We translated the WEEE quantities into secondary resourcesvailable for recovery and the associated economic potential. Fig. 6hows the amount of key resources over time, presented in fourroups: 1 – common materials (plastics and base metals (Al, Cu,e)), 2 – common metals (Pb, Sn, Zn), 3 – less common metals (Ba,i, Co, Ga, Sr, Ta), and 4 – precious metals (Ag, Au, Pd).

The amounts of common materials follow the similar patternompared to that of WEEE over time. For example, the amounts ofe and plastic have increased from 21 and 16 kt in 1990 to 38 and0 kt in 2015, respectively. These amounts will increase slowly upo 40 and 31 kt by 2025. Other resources, including precious met-ls show different patterns of increase, reaching their peak around013 followed by a smooth decline. The total amount of Au, forxample, increased from 0.6 t in 1990 to 1.04 t in 2013 and pro-



ected to go down to 0.97 t in 2025. Similarly, the amounts of Sn,a, Ta, and Ag decrease rapidly after the peak. The trends show thathe base materials, which are mainly used in the structural com-onents of the products, remain largely unchanged. On the other

PRESS and Recycling xxx (2016) xxx–xxx 5

hand, the specialty and precious metals, which are used in traceamounts in the products – usually for achieving specific function-ality – fluctuate more. This change in material composition of WEEEcould be attributed the change in the composition of products typesover time.

The potential revenue carried by the secondary resources avail-able in WEEE is shown in Fig. 7. The total revenue increased fromD 69 million in 1990 to D 122 million in 2015. Au and plastic are thetwo key resources, carrying more than half of the total revenue. Cuis the next important metal that represents more than 15% of thepotential revenue, while remaining is made up largely by Fe andBa. However, the actual exploitable revenues could significantlyvary among different resources depending on the technological andeconomic feasibility of their recovery. Further, the actual materialrecovery also depends on the type of products and their EoL man-agement system. For example, precious metals (e.g. Au and Pd) areconcentrated in printed circuit boards (PCBs), therefore, selectivedismantling and separate processing of PCBs are preferable for theirefficient recovery (Ardente et al., 2014). In a typical setup supportedby such dismantling process, precious metals as well base metals(Al, Cu and Fe) can be effectively recovered by the existing recyclingsystem, whereas plastics and less common metals (including rareearth elements) have considerably low recycling rates (Habib et al.,2015; Parajuly et al., 2016).

3.3. WEEE management in Denmark

3.3.1. CollectionDenmark has been performing consistently well with respect to

the collection target (i.e., 4 kg per capita per year) set by the WEEEDirective. However, the recast of the Directive requires:

From 2019, the minimum collection rate to be achieved annuallyshall be 65% of the average weight of EEE placed on the marketin the three preceding years in the Member State concerned,or alternatively 85% of WEEE generated on the territory of thatMember State.

We compared the trend of existing collection with the estimatedamount of WEEE generation in the future and the new targets. Asshown in Fig. 8, the collection for the years 2006–2013 is close tothe 65% target. According to the Directive (recast), 85% of WEEEgenerated is broadly equivalent to 65% of the PoM EEE averagedover three preceding years. However, our estimates suggest a sig-nificant difference between these two numbers. This means thereis a room for error, and the targets are not equally applicable forcountries like Denmark.

It is important to understand not only the gap between PoMEEE and collected WEEE quantities, but also what is actually beingcollected. The numbers reported by DPA (Fig. S1 in SM) show a sig-nificant gap between PoM EEE and collected WEEE for the collectionfraction ‘small appliances’, which covers all household productsof categories SHA, ITE and CE. It shows that not all EoL productscoming out from the households enter the official WEEE collectionsystem. On average, 64% of overall PoM EEE (the same year) wascollected between 2006 and 2014, but this average is only 40% forthe fraction ‘small appliances’.

3.3.2. RecyclingAs reported by the DPA-system, most of the collected WEEE in

Denmark goes for material and energy recovery. For the four cat-egories, 74% to 88% of recycling rates have been reported for the


(recast, Article 11 (2)):

The achievement of the targets shall be calculated, for eachcategory, by dividing the weight of the WEEE that enters the




Fig. 6. Amount of resources available for recycling in the EoL household products.

resou

n

R

n

Fig. 7. Potential revenue of the secondary

recovery or recycling/preparing for re-use facility [. . ..] by theweight of all separately collected WEEE for each category,expressed as a percentage.

This can be expressed using the flows from Fig. 2 (assuming theet trade of EoL products to be zero):



ecycling rate (Directive target) = A7−8 + A7−4

A5−7. (3)

However, looking at the whole system level, this indicator doesot reflect the true material recycling rate as it excludes the com-

rces in the estimated quantities of WEEE.

plementary flow of WEEE (A6–7) from the equation. Moreover, itdoes not consider the losses that occur in the WEEE treatment chain(A8and A10) or the EoL products handled outside the official recy-cling routes (A7). In order to reflect the true material recovery, thecalculation should include the remaining recycling routes in theequation.


The targets set by the Directive (recast) also include reuse of EoLproducts. However, the above equation does not include the prod-ucts entering reuse market without coming to the official WEEEcollection system. A significant amount of used and EoL products



K. Parajuly et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx 7

F cenarc

(wro–Acotm

bMtiep

3

neipta

efiippanafli

am

ig. 8. Illustration of products sales (PoM EEE), WEEE generation, and collection sollection targets defined by the WEEE Directive (recast).

especially ITE and CE) enter their second life via commercial asell as user-to-user selling platforms. Further, reuse needs a sepa-

ate evaluation, as it can also be found within the system boundaryf ‘in-use’. Only the WEEE – excluding the EoL products for reuse

should be used for calculating the collection and recycling rates.s it can be seen, the metrics used for calculating recycling ratesan be influenced by the different flows in the WEEE system. Inrder to make the recycling rates comparable across different coun-ries (or regions), it is important to understand all flows in a WEEE

anagement system and include them in a consistent manner.Efficient recovery of secondary resources allows environmental

enefits that come from the avoided mining of these resources.oreover, resource recovery from EoL products helps reducing

he potential supply risks of important resources – an issue ofncreasing concern for industries and governments alike. Therefore,fficient collection and recycling of WEEE needs to be prioritized atolicy level.

.3.3. Complementary flowsToday, the EoL products collected outside the official system are

ot included in the equation of calculating recycling and recov-ry rates. Exclusion of such flows not only affects the performancendicators of the WEEE management system, but also lessens theossibilities of recovering more value from the EoL products. Therue recycling achievements should be compared with what isvailable, and not only with the amount that is already collected.

The amounts of EoL products (flow A3–6 in Fig. 2) that do notnter the official WEEE management system have not been quanti-ed. The amount of stockpiled products, both during and after use

n the households, is also unknown. Considerable amounts of usedroducts are diverted from the WEEE flows through user-to-userlatforms, stores that repair and resale the used items, or as anct of charity. These flows (represented by the flow A7–4 in Fig. 2)eed to be quantified and the delay caused by the reuse of productslso needs to be addressed. Quantification of these complementaryows and their inclusion in the metrics will make the recycling



ndicators more robust and precise.In order to keep track of the resource flows more precisely

nd transparently, harmonization of data sources and reportingethodology is also crucial. The sales data available (from both

ios. The lines ‘65% of PoM EEE (3-year average)’ and ‘85% of WEEE’ represent the

Statistics Denmark and Euromonitor) for EEE give us the number(or count) of marketed products, whereas the quantity reportedby the producers (to the DPA-System) is the weight of PoM EEE.The information on how the number of products is translated intoweight is not available. This suggests a lack of clarity among differ-ent stakeholders on their role and format of reporting data. Thereis also a lack of documentation for products imported or exportedat personal levels, and the household stocks of products in use,hibernating and/or stockpiled after being obsolete. More bottom-up analysis and consumer surveys are therefore needed.

3.4. Limitations and uncertainties

Although we provide a comprehensive view of the products’ lifecycle and quantify the flow using dynamic MFA, we acknowledgea few limitations and uncertainties in our analysis.

The first limitation has been the number of products includedin this study. Due to the lack of sales data, we considered only61 product types from four categories, which leaves out other EEEitems available in market and the WEEE stream. Nevertheless, the61 product types cover on average 80% (by weight) of the total PoMcompared to what is reported by DPA for the period of 2006–2014.This shows that the products we considered sufficiently representthe household EEE and the corresponding WEEE. Considering theincreasing number of EEE items entering the market every year,it will remain a challenge to include the whole spectra of producttypes for a given time window.

The weight and material composition of products are key vari-ables for this study but they bear some degree of uncertainties. Weconsidered fixed values for these product characteristics at cat-egory level, but in reality, they may change with the continuingtechnological evolution. The change in the material composition isespecially crucial because the content of precious metals may havebeen decreasing. The recent decades have seen such an evolution,especially in the IT equipment and consumer electronic productssuch as computers, tablets, music players, televisions, and mobile


phones. These factors play an important role in the availability ofEoL products and the secondary resource in WEEE.

The lifespan of products are the key for dynamic MFA stud-ies; however, we have very little information on product lifespans


ING ModelR

8 vation

b2aBif

fapietpfal

paedldpa

4

iassahei

A

tTptp

A

i0

R

A

B

classification of WEEE. Electron. Goes Green 2012+ (EGG) 2012, 1–6.Zeng, X., Gong, R., Chen, W.Q., Li, J., 2015. Uncovering the recycling potential of

‘new’ WEEE in China. Environ. Sci. Technol.Zhang, L., Yuan, Z., Bi, J., Huang, L., 2012. Estimating future generation of obsolete

household appliances in China. Waste Manage. Res. 30, 1160–1168.

ARTICLEECYCL-3325; No. of Pages 8

K. Parajuly et al. / Resources, Conser

ased on firsthand statistics (Liu and Muller, 2013; Oguchi et al.,010). Due to the lack of specific Danish data, product lifespan isdapted from a study conducted for the Netherlands, France, andelgium. The lifespan may also change overtime (e.g., due to chang-

ng consumer behavior and product design), which is an importantactor to consider in future WEEE MFA studies.

We counted all the future generation of secondary resourcesor potential revenues. In reality, of course, not all the recyclingnd recovery would be technically possible due to, for exam-le, contamination and accumulation of alloying elements. More

mportantly, the resource recovery is hampered when the recov-ry of valuable elements itself is not economically feasible and/orhe total recycling cost – including the proper treatment and dis-osal of hazardous substances – is too high. The expected revenues

rom the resources in WEEE can also be influenced due to fluctu-ting market conditions, the grade of resource, and geographicalocation.

We did not include a quantitative uncertainty analysis for thesearameters in this study because there is little information avail-ble on their historical and potential future trend. Moreover, thexpected revenues from the resources in WEEE can be influencedue to fluctuating market conditions. Given the diversity and evo-

ution of EEE items, it will require significant efforts to reflect theynamic nature of product characteristics and their EoL recoveryotential. We see this as a common challenge that both the WEEEnd the MFA research community should take in the future.

. Conclusions

The study affords a detailed understanding of the WEEE flowsn Denmark – both qualitatively and quantitatively. The system-tic quantification of the WEEE and the secondary resources isupported by a comprehensive analysis of the WEEE managementystem. Although the performance of the system has been sufficientgainst current WEEE Directive, new set of legislations will requireigher collection and recycling rates. A more robust and transpar-nt documentation of the flows will help achieve these targets andmprove the overall resource recovery from WEEE.

cknowledgements

This research was a part of the INNOSORT project (innosort.eknologisk.dk) that is funded by the Danish Agency for Science,echnology and Innovation. We thank the Agency and the projectartners for their support. The views contained in the paper arehose of the authors and do not represent the official views orolicies of any stakeholders.

ppendix A. Supplementary data

Supplementary data associated with this article can be found,n the online version, at http://dx.doi.org/10.1016/j.resconrec.2016.8.004.

eferences

rdente, F., Mathieux, F., Recchioni, M., 2014. Recycling of electronic displays:analysis of pre-processing and potential ecodesign improvements. Resour.



Conserv. Recycl. 92, 158–171.alde, C.P., Kuehr, R., Blumenthal, K., Fondeur Gill, S., Kern, M., Micheli, P.,

Magpantay, E., Huisman, J., 2015a. E-Waste Statistics: Guidelines onClassifications, Reporting and Indicators. United Nations University, IAS –SCYCLE, Bonn, Germany.

PRESS and Recycling xxx (2016) xxx–xxx

Balde, C.P., Wang, F., Kuehr, R., Huisman, J., 2015b. The Global E-Waste Monitor –2014. United Nations University, IAS – SCYCLE, Bonn, Germany.

Bigum, M., Petersen, C., Christensen, T.H., Scheutz, C., 2013. WEEE and portablebatteries in residual household waste: quantification and characterisation ofmisplaced waste. Waste Manage. 33, 2372–2380.

Brunner, P.H., Rechberger, H., 2004. Practical Handbook of Material Flow Analysis.Springer.

Cucchiella, F., D’Adamo, I., Lenny Koh, S.C., Rosa, P., 2015. Recycling of WEEEs: aneconomic assessment of present and future e-waste streams. RenewableSustainable Energy Rev. 51, 263–272.

DPA-System, 2015. WEEE, BAT, and ELV Statistics Denmark. DanskProducentansvars System.

de Souza, R.G., Climaco, J.C., Sant’Anna, A.P., Rocha, T.B., do Valle, R.A., Quelhas, O.L.,2016. Sustainability assessment and prioritisation of e-waste managementoptions in Brazil. Waste Manage.

Duygan, M., Meylan, G., 2015. Strategic management of WEEE inSwitzerland—combining material flow analysis with structural analysis.Resour. Conserv. Recycl. 103, 98–109.

Elshkaki, A., 2005. Dynamic stock modelling: a method for the identification andestimation of future waste streams and emissions based on past productionand product stock characteristics*1. Energy 30, 1353–1363.

Euromonitor International, 2016 https://www.portal.euromonitor.com.European Parliament, 2003. Directive 2002/96/EU of the European Parliament and

of the Council of 27 January 2003 on waste electrical and electronic equipment(WEEE). Off. J. Eur. Union.

European Parliament, 2012. Directive 2012/19/EU of the European Parliament andof the Council of 4 July 2012 on waste electrical and electronic equipment(WEEE) (recast). Off. J. Eur. Union.

Eurostat, 2016. Waste Electrical and Electronic Equipment (WEEE).FEHA, 2016. Danish Association for Suppliers of Electrical Domestic Appliances.Habib, K., Schibye, P.K., Vestbo, A.P., Dall, O., Wenzel, H., 2014. Material flow

analysis of NdFeB magnets for Denmark: a comprehensive waste flowsampling and analysis approach. Environ. Sci. Technol. 48, 12229–12237.

Habib, K., Parajuly, K., Wenzel, H., 2015. Tracking the flow of resources inelectronic waste – the case of end-of-life computer hard disk drives. Environ.Sci. Technol. 49, 12441–12449.

Kalmykova, Y., Patricio, J., Rosado, L., Berg, P.E., 2015. Out with the old, out with thenew—the effect of transitions in TVs and monitors technology on consumptionand WEEE generation in Sweden 1996–2014. Waste Manage. 46, 511–522.

Liu, G., Muller, D.B., 2013. Centennial evolution of aluminum in-use stocks on ouraluminized planet. Environ. Sci. Technol. 47, 4882–4888.

Mudd, G.M., 2010. The Environmental sustainability of mining in Australia: keymega-trends and looming constraints. Resour. Policy 35, 98–115.

Muller, E., Hilty, L.M., Widmer, R., Schluep, M., Faulstich, M., 2014. Modeling metalstocks and flows: a review of dynamic material flow analysis methods.Environ. Sci. Technol. 48, 2102–2113.

Oguchi, M., Murakami, S., Tasaki, T., Daigo, I., Hashimoto, S., 2010. Lifespan ofcommodities, part II. J. Ind. Ecol. 14, 613–626.

Oguchi, M., Murakami, S., Sakanakura, H., Kida, A., Kameya, T., 2011. A preliminarycategorization of end-of-life electrical and electronic equipment as secondarymetal resources. Waste Manag. 31, 2150–2160.

Parajuly, K., Habib, K., Cimpan, C., Liu, G., Wenzel, H., 2016. End-of-life resourcerecovery from emerging electronic products—a case study of robotic vacuumcleaners. J. Clean. Prod. 137, 652–666.

Polak, M., Drapalova, L., 2012. Estimation of end of life mobile phones generation:the case study of the Czech Republic. Waste Manage. 32, 1583–1591.

Sharma, V., Garg, S.K., Sharma, P.B., 2016. Identification of major drivers androadblocks for remanufacturing in India. J. Clean. Prod. 112, 1882–1892.

Statistics Denmark, 2016. The families possession of home appliances by type ofconsumption.

Wang, F., Huisman, J., Balde, K., Stevels, A., 2012. A systematic and compatible



http://innosort.teknologisk.dk



http://dx.doi.org/10.1016/j.resconrec.2016.08.004











http://refhub.elsevier.com/S0921-3449(16)30196-3/sbref0005





























































































































































http://https://www.portal.euromonitor.com







































































































































































































































































































































g model article in press - uzh 144/2016_parajuly... · discarded weee in residual waste from danish...

Documents